http://www.amazon.com/Beyond-Oil-View-Hubberts-Peak/dp/0809029561

Over a period of time, however, the pressure tails off and it becomes increasingly difficult to extract the oil. Gas injection is then used to recreate / maintain enough pressure to continue pumping. This gas is injected below the reservoir (as Tim Newman pointed out) and succeeds in forcing many more of those tiny droplets towards the surface.

It’s a well-established technology that’s been in use for many many years (many of the oil wells I’ve visited have been using some form of gas injection) and while, ideologically, I believe we should be leaving fossil fuels in the ground right now; there’s no question that the technology does indeed work and has kept hundreds of US (and other) wells pumping long after they would otherwise have ceased production.

Not necessarily. BP could get the oil out without reinjection, but the production rates would not justify the costs of extraction. With gas reinjection, it would have. BP’s likely option now they’re not reinjecting will be to walk away, flogging the field to a much smaller independent operator (which are flourishing in the North Sea now) whose extraction costs will be much less than those of BP.

Oil recovery by gas reinjection works by injecting the gas beneath the oil reservoir. The pressure buildup below the oil then forces the oil up through the well.

Well, I’ve learnt that the hard way, Kay Tie. My reasoning — now well-established as faulty, no need to rub any more salt in the wound, folks — was simply an assumption that the press-release would contain their most optimistic projections and claims, and thus the best ones to challenge.

As I’ve said however… that was clearly a mistake.

Water is to hand (offshore, anyway) and neatly doesn’t compress. CO2 is a pain in the butt. And toothpaste is really a pain in the butt since the purchasing department can’t spell “fluoride” and thus none ever gets delivered.

I’m sure they thought of this before they invested millions in the idea. While the Government wouldn’t hesitate to waste millions on a doomed-from-the-start idea, BP wouldn’t be quite so wasteful of their own money.

Assuming that CO2 can be compressed to liquid form to the same density as oil, and displaces the same volume of oil, which is mainly-carbon-with-a-bit-of-hydrogen; and one unit of oil burns to form 3.7 times its weight in CO2, I fail to see how this can lead to an overall reduction in CO2.

Never take a press release at face value. It’s usually written by a (junior) marketing droids who knows nothing about the topic. A classic example was the press release from NASA JPL that had converted the dimensions of the Mars Orbiter into metres by multiplying feet by 3.28. (ironic considering the 1998 failure of the orbit due to a software fault in converting between imperial and metric units).

I’d like to point out though, that I still feel your initial post on this subject was a cheap shot and I’ll be responding to it at some point.

If Professor Blunt’s figures are correct, then of course my calculations need to be revised.

However, his figures are do not tally with those stated by BP.

“The project would also permanently store 1.3 million tonnes of carbon dioxide, the equivalent of removing 300,000 cars from the roads.”
From: BP website

BP do not claim that the figure is “per annum”, and indeed the context would appear to imply that it is not.

Once again, though, if BP have themselves under-reported the CO2 being captured then my calculations clearly require revision. But I’m not sure I can really be held responsible for inaccurate data provided by those planning the project.

Tim adds: A Google search for “BP Peterhead” gives as result two or three a page which says this:

“When fully operational, the project would be expected to capture and store around 1.3 million tonnes of carbon dioxide each year and provide ‘carbon-free’ electricity to the equivalent of a quarter of a million UK homes.”

http://www.bp.com/genericarticle.do?categoryId=2012968&contentId=7006999

So per annum I think?